This R21 project addresses the NCI's Provocative Questions #PQB2. The significance of this proposal will be to establish principles governing the initial immune recognition of malignancy, and how the immunosuppressive tumor microenvironment emerges. A fundamental problem in cancer biology is that we do not understand the rules governing the immune-tumor interplay, nor do we understand how the immune response recognizes and interacts with the early phases of an emerging malignancy. Whether the immune system constrains cancer throughout life but occasionally fails remains contentious. A related issue of contention concerns the mechanisms of how cancer is discriminated from self: attempts to address this question have spurned decades of research into tumor antigens and their potential clinical uses. Macrophages are the earliest immune cells to populate a nascent malignancy. Why macrophages populate tumors remains unexplained, but is a probably a consequence of their tissue repair and recovery functions: macrophages are likely surveying an 'organ' for health. Macrophages come from different sources: the yolk sac/fetal liver-derived tissue macrophages and macrophages originating from the blood monocyte pool. In the earliest phases of tumor growth, the source of macrophages is unclear. Our model is the first steps of malignancy involve an interplay between macrophages from both tissue and blood sources conspiring to establish the tissue repair and immunosuppressive microenvironment. Therefore, experimental dissection of the initial recognition of tumors by myeloid cells is one key to understanding how malignancy is established. To dissect the initial immune-tumor interplay we will use a developmental model of cancer: Neuroblastoma (NB), the most common solid tumor of childhood. Through integrated genetic, imaging and immunological approaches we will trace the ontogeny of immune recognition of early malignancy without the need to orthotopically transplant tumor cells to initiate cancer. To address PQB2 we will determine the principles governing myeloid recruitment to a growing tumor across time (Aim 1). Using genetic models of NB, a developmental tumor growing in a defined location and within a predictable time window, we will quantify the source and properties of myeloid cells infiltrating tumors from the earliest detectable size to larger malignancies. In Aim 2 we will quantify the pro- and anti-tumor T cells in nascent malignancies. Using NB, we will quantify CD8+ T cell responses using tetramer staining for a tumor linked antigen and correlate these data with the number and activity of local Treg. By combining tetramer staining, in vivo visualization techniques and manipulation of tissue versus monocyte-derived macrophages we will determine how T cells interact with growing malignancies, and the relationship between the T cell response and local myeloid populations.